JOURNAL O F T H E A M E R I C A N C H E M I C A L SOCIETY Regislered in U. S. Palcnl Ofice.
@ Cogyrighl
1971 b y lhe American Chemical Sociely
MAY5, 1971
VOLUME93, NUMBER 9
Physical und Inorgunic Chemistry Mechanisms of Chemiluminescent Electron-Transfer Reactions. I. The Role of the Triplet State in Energy-Deficient Systems D. J. Freed and Larry R. Faulkner*
Contributionfrom the Coolidge Chemical Laboratory, Harvard University, Cambridge, Jkfassachusetts 02138. Received July 17, 1970 Abstract: Two separate studies were carried out to elucidate the role of the triplet state in energy-deficientchemiluminescent electron-transfer reactions occurring in N,N-dimethylformamide. The first of these was an investigation of the nature of luminescence from a series of reactions involving the radical anions of anthracene, 9,lO-diphenylanthracene, rubrene, and fluoranthene as reductants and the radical cations of N,N-dimethyl-p-phenylenediamine and 10-methylphenothiazineas oxidants. All experiments were conducted with reactants generated at an electrode at controlled potential. The results for systems containing fluoranthene indicate that a critical value of electron-transfer enthalpy exists which must be exceeded for luminescence to occur. The value is near the energy required to excite fluoranthene to its lowest triplet state. A second investigation was a study of the effect of a series of chemically inert triplet quenchers on the reaction between the fluoranthene radical anion and the 10-methylphenothiazine radical cation. trans-Stilbene, anthracene, and pyrene were able to quench the normal fluoranthene emission. Moreover, the systems containing anthracene and pyrene emitted light characteristic of the fluorescence of these molecules. This behavior was interpreted as providing prima facie evidence for a required triplet intermediate in the energy-deficient chemiluminescent reactions of fluoranthrene. The results are discussed in terms of possible mechanisms for luminescence.
B
ecause electron-transfer reactions are fundamental to all of chemistry, and because the chemiluminescent electron-transfer reactions involving radical ions of aromatic compounds offer a unique opportunity to study the detailed course of one type of electron transfer, a significant effort has been invested recently in mechanistic studies of these systems. l v 2 The simpler systems feature only one emission component, which is spectrally identical with the fluorescence of one of the products of the electron-transfer reaction. In these systems it is clear that the emitting first excited singlet state is populated in some way as an accompaniment to the electron-transfer process. Indeed, most current thinking presumes a direct population of some excited state of a product molecule in this process. Clearly the relative energetics of excitation and electron transfer are important considerations in such an hypothesis. As a consequence, one of the most intriguing aspects of (1) A. Zweig, Aduan. Photochem., 6, 425 (1968). (2) D. M. Hercules, Accounts Chem. Res., 2, 301 (1969).
the mechanistic studies has been the problem of “energy-deficient systems,” i.e., those for which an emitted photon contains far more energy than is released by a single electron-transfer event. 334 Thus, energy considerations for such cases apparently rule out direct production of the emitting state during electron transfer. One must rely on some energy-multiplying step to account for the observed behavior in terms of excitedstate generation during the redox process. Most speculation to date has employed the hypothesis that the redox reactions generate a triplet-state product molecule, which in turn produces the emitting singlet in a following triplet-triplet annihilation. An alternate mechanism, advanced by Zweig, et al., uses the hypothesis that the emitting state is generated directly by multiple electron transfer to or from aggregates of radical ions.5 This latter view envisions the energy lib(3) A. Weller and K. Zachariasse, J. Chem. Phys., 46,4984 (1967). (4) A. J. Bard and K. S.V. Santhanam, J . Amer. Chem. SOC., 87,139 (1965).
2097
2098
l l I /
-REFERENCE
ELECTRODE
JOINT
one study of a group of systems involving only one possible emitter, we sought to determine whether one could observe emission accompanying electron-transfer reactions which are not energetic enough to populate even the emitter triplet state directly. Thus we sought to discover whether a “critical enthalpy” exists for the emission of light, and whether that enthalpy is related to the lowest triplet level of the emitter. In a second, study, we investigated the effects of suitable triplet quenchers on the emission from chemiluminescent systems. Because fluoranthene has a high triplet energy compared to its reduction potential, it proved to be an eminently suitable substrate for this work. We believe that the results reported here represent the first hard experimental evidence that a triplet state is a necessary intermediate for chemiluminescence from energy-deficient systems.
Experimental Section
c- COUNTER fLECTROOE
3mm OD WORKING ELECTRODE
I
1
Figure 1. Electrolysis cell used for generation of radical ions.
erated in such a step as being localized on one molecule, a constituent of the former aggregate, which then emits. Experimental results bearing on these hypotheses have been elusive, and there are few informative reports available. In a study of several energy-deficient reactions using the Wurster’s Blue radical cation as an oxidant, Weller and Zachariasse have reported that the emission resulting from chrysene radical anion oxidation contains two componentsma One of these corresponds to the expected chrysene fluorescence band. The second matches the chrysene phosphorescence spectrum and indicates the presence of the chrysene triplet state. In other work, Chang, Hercules, and Roe have shown that a series of double-potential-step experiments yields behavior which is kinetically consistent with the triplet mechanism if one assumes the radical ions to be effective quenchers of triplet statesS6 Finally, Faulkner and Bard studied magnetic field effects on chemiluminescence and delayed fl~orescence.~They concluded that these effects can be rationalized in terms of the triplet mechanism, and especially in terms of radical-ion quenching of the triplet intermediates, which they actually found to be an efficient process. Because it offers the simplest available explanation, and because of these probing experiments, the triplet mechanism is the more widely favored one. Nonetheless, no published report demonstrates that a triplet is a precursor to the emitting singlet. In an effort to fill this void, we have carried out experiments bearing on the problem from two separate approaches. In ( 5 ) A. Zweig, A. K. Hoffman, D. L. Maricle, and A. H. Maurer, J . Amer. Chem. Soc., 90, 261 (1968). (6) J. Chang, D. M. Hercules, and D. I
